David Hume and Deriving an “Ought” from an “Is”

It seems easy to make an ethical argument against punching someone in the face. If you do it, you will physically harm the person. Therefore, you shouldn’t.

But the 18th century philosopher David Hume famously argued that inferences of this type – in which what we ought morally to do (not punch someone) is derived from non-moral states of affairs (punching him will hurt him) – are logically flawed. You cannot, according to Hume, derive an “ought” from an “is,” at least without a supporting “ought” premise. So, deciding that you ought not punch someone because it would harm him presupposes that causing harm is bad or immoral. This presupposition is good enough for most people.

But for Hume and those who subscribe to what is now commonly referred to as the “is-ought gap” or “Hume’s guillotine,” it is not enough.

Hume put the heads of preceding moral philosophers in his proverbial guillotine in Book III, Part I, Section I of his A Treatise of Human Nature. He wrote that every work of moral philosophy he had encountered proceeded from factual, non-moral observations about the world to moral conclusions – those that express what we ought or ought not do. The shift is imperceptible, but it is a significant blunder. “For as this ought, or ought not, expresses some new relation or affirmation, it is necessary that it should be observed and explained; and at the same time that a reason should be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it.”

The blunder, according to Hume, is one of logic. Factual statements are logically different from moral statements, so no factual statements can, by themselves, entail what people morally ought to do. The “ought” statement expresses a new relation, to use Hume’s phrase, that isn’t supported by its purely factual premises. So, a moral judgment that is arrived at by way of facts alone is suspect.

The new, unexplained relation between moral judgments and solely factual premises is characteristic of the broader distinction between facts and values. Moral judgments are value judgments – not factual ones.

In the same way, judgments about the tastiness of particular foods are value judgments. And positive and negative assessments of foods are not logically entailed by just the facts about the foods.

If a cheese enthusiast describes all the facts he knows about cheese – like that it’s made from milk, cultured bacteria, and rennet – that wouldn’t be enough to convince someone that its delicious. Nor would a cheese hater’s listing all the same facts prove cheese is disgusting. Both the cheese lover and the cheese hater make an evaluation that isn’t governed strictly by the logical relations between the facts.

Despite the logical gap between “is” and “ought” statements, and the broader distinction between facts and values, Hume didn’t think moral judgments are hogwash. He just thought they come from sentiments or feelings rather than logical deductions. In Book III, Part I, Section II of the Treatise, he wrote, “Morality … is more properly felt than judged of; though this feeling or sentiment is commonly so soft and gentle, that we are apt to confound it with an idea, according to our common custom of taking all things for the same, which have any near resemblance to each other.”

So, Hume would most likely agree that punching someone in the face is wrong. But he’d say an argument against it is unnecessary, even mistaken. People feel the wrongness. They feel that one ought not punch another in the face – just like a punched person feels the pain.

Do Moral Facts Exist?

Virtually all non-psychopaths think murder is morally wrong. But what makes it so? Is the wrongness an objective fact, one that would exist no matter how people felt about it? Or does the wrongness of murder reside only in people’s minds, with no footing in objective reality?

The question falls in the branch of moral philosophy called metaethics. Instead of pondering topics that come up during everyday moral debate – such as whether a given action is right or wrong – metaethics is more abstract. It is concerned with the nature of morality itself. What do people mean when they say something is right or wrong? Where do moral values come from? When people make moral judgments, are they talking about objective facts or are they merely expressing their preferences?

So, the objectivity of murder’s wrongness depends on whether objective moral facts exist at all. And not all moral philosophers agree that they do.

On one side are the moral realists, who say there are moral facts and that these facts make people’s moral judgments either true or false. If it is a fact that murder is wrong, then a statement that it’s wrong would be true in the same way that saying the earth revolves around the sun would be true.

Moral antirealists hold the opposite. They say there are no moral facts and that moral judgments can’t be true or false like other judgments can.

Some argue that when people express moral judgements, they aren’t even intending to make a true statement about an action. They are simply expressing their disapproval. The philosopher A.J. Ayer popularized this perspective in his 1936 book Language, Truth, and Logic. He argued that when someone says, “Stealing money is wrong,” the predicate “is wrong” adds no factual content to the statement. Rather, it’s as if the person said, “Stealing money!!” with a tone of voice expressing disapproval.

Because moral statements are simply expressions of condemnation, Ayer said, there is no way to resolve moral disputes. “For in saying that a certain type of action is right or wrong, I am not making any factual statement . . . I am merely expressing moral sentiments. And the man who is ostensibly contradicting me is merely expressing his moral sentiments. So that there is plainly no sense in asking which of us is right.”

Other antirealists are on the realists’ side in thinking that moral discourse makes sense only if it assumes there actually are moral facts. But these antirealists – called “error theorists” – say the assumption is false. People do judge actions to be right or wrong in light of supposed moral facts, but they are mistaken – no moral facts exist. Thinking and acting as if they do is an error.

Moral Realism graphic

“The strongest argument for antirealism,” says Geoffrey Sayre-McCord, a philosopher at the University of North Carolina at Chapel Hill, “is to point out the difficulty of making good sense of what the moral facts would be like and how we would go about learning them.”

Scientists can peer through microscopes to learn facts about amoebas. Journalists can observe press conferences and report what was said. Toddlers can tell you that the animal on the sofa is a brown dog. The job of the moral realist is to show that there are moral facts on par with these readily-accepted types of non-moral facts.

Sayre-McCord, who considers himself a moral realist, says this is done best by thinking about what would have to be true for our moral thoughts to be true. This results in some sophisticated philosophical accounts, he says.

Justin McBrayer, a philosopher at Fort Lewis College, says the truth of moral claims can be evaluated by analogy to the ways non-moral truths are established. The same “epistemic norms” apply whether a moral claim or a non-moral claim is being defended. “Some arguments are good, and some are bad,” he says.

Most philosophers are moral realists, but there is a sizeable minority in the antirealist camp. In a 2009 survey of professional, PhD-level philosophers, 56% said they accepted or leaned toward moral realism, while 28% said they accepted or leaned toward moral anti-realism. Sixteen percent said they held some other position.

McBrayer and Sayre-McCord point out the lack of data on the general population’s views, but they both sense that the default position among non-philosophers is moral realism. People think, act, and speak as if there are objective moral facts. But since most have never considered the alternative, many have trouble when pressed to defend their views. “They have to stop and think about it,” McBrayer says.

Sayre-McCord says most people tend to back away from their commitment to moral realism when they’re challenged. “There is a tendency for people to be antirealists metaethically, but realists in practice.”

There is no doubt that how people think about morality affects their behavior, McBrayer says. Psychological research backs this up. In one study, researchers found that participants “primed” to think in realist terms were twice as likely to donate to a charity than participants primed to think in antirealist terms. In another study, researchers found that participants who read an antirealist argument were more likely to cheat in a raffle than those who read a realist argument.

Given these findings, even if murder’s wrongness is just a fiction, it’s hard to argue that it’s not a useful one.